You can finagle a situation where you're using a console preamp not routed to the mix, but just routed to a direct out that feeds Smaart. It's very important to make sure that that input's entire path through the console is devoid of high pass filters, compressors, gates, or other things that might be on by default. This is helpful in applications where I'm using the console's matrices as some form of loudspeaker processing.
What I WILL do with some frequency is run the output of Smaart's signal generator through the console, and from there feed it to the PA. You can do what you suggest - and I have, although I generally avoid it because it affords too many opportunities for crossed wires / disasters (ever put a measurement mic through the PA by accident? It's a mistake you only make once). The question to ask yourself is what are you trying to measure? In other words, do you want the console's response to show up in the measurement, or not? (The pragmatic answer is "both", which is why in the SOFO sessions, Jamie has a Console transfer function measurement that measures the console's output against its input, and then a number of other transfer function engines that measure downstream from that point, using the console output as a reference. Glad you are thinking critically about this, because it's one of the most important decisions we have to make when we're preparing to take a measurement. I'm still new to this world, so please let me know if my thought process is off in any way. Wouldn't this be cancelled out in the TF results? signal and the measurement signal - assuming all channel processing other than gain is turned off. Would there be a reason as to not run the mics through the mixers preamps? From what I understand, this would in general be the same processing to both the ref. This does mean the DSP from the mixers preamp is applies to the ref. The "norm" is getting the reference signal after the mixer, so we don't take the internal DSP into consideration in our measurements. This got me thinking, why don't I use the mixer instead of a dedicated card. I almost always carry my mixer with me to gigs, or have a dante/ otherwise connected mixer available. I have a Scarlet 2i2 IO card for simple gigs, but this only gives me inputs for 1 TF at a time, which isn't always ideal. The magnitude of my TF setups varies widely from day to day and gig to gig, as I imagine it does for most people.
I've run through all but 1 of the SOFO webinars, but as far as I can tell this was only briefly mentioned. So, this is something I've thought about quite a bit lately.